A Comparative Econometric Analysis of Rice and Banana Production and Household Risk Responses under Riverbank Erosion in Pabna District, Bangladesh

Abstract

Riverbank erosion along the Padma and Jamuna rivers remains a major constraint on agricultural production and rural livelihood resilience in Bangladesh. This study compares the production performance, resource-use efficiency, profitability, and erosion-related monetary gaps of rice and banana cultivation and examines household risk responses in Pabna District. The analysis is based on primary cross-sectional data collected in August 2025 from 300 farm households, equally divided between erosion-prone and non-erosion areas. Separate Cobb–Douglas production functions were estimated for rice and banana in the two environmental settings. Resource-use efficiency, profitability indicators, monetized return gaps, and household risk-perception measures were also analysed. The results reveal substantial crop- and location-specific differences. In erosion-prone areas, only seed cost is marginally significant in the rice and banana models, while the remaining inputs show weak gross-return responsiveness. The sums of the estimated elasticities are 0.168 for rice and 0.204 for banana, indicating decreasing returns to scale. In non-erosion areas, seed, pesticide, fertilizer, labour, and land rent are positively and significantly associated with gross return for both crops. The corresponding returns-to-scale estimates are 2.51 for rice and 2.88 for banana, indicating increasing returns to scale. Rice cultivation generates a benefit–cost ratio of 1.21 in erosion-prone areas and 1.98 in non-erosion areas, while the corresponding ratios for banana are 1.07 and 2.22. The comparative gross-return gaps are Tk. 30,000 per bigha for rice and Tk. 66,300 for banana, while the net-return gaps are Tk. 25,000 and Tk. 62,300, respectively. Household-level evidence further indicates high perceived vulnerability and risk-averse responses, including preference for short-duration crops, cautious investment, income diversification, and reliance on community-based coping mechanisms. The findings demonstrate that riverbank erosion is associated with weak input responsiveness, inefficient resource allocation, low profitability, and constrained agricultural investment. Integrated interventions combining riverbank protection, crop-specific extension, risk-sensitive credit, crop insurance, social protection, and livelihood diversification are required to strengthen resilience in erosion-prone communities.